• Thom A - Tuesday, April 17, 2018 4:37 AM

    Personally, i think Phil's suggestion earlier is the winner. Check if the rogue column exists first, if it doesn't use the dataflow that doesn't require it, if it does, use the one that does have it.

    That will require me to have to separated data flows with the same code. So not so easy to maintain.

    I would prefer to have just one Data Flow with two different data source flat files inside.

    This is way I would be very pleased if I understand what can I do to guaranty:

    Both data sources connected to the same Union all. If one fails, the other should continue the process without problem...   Problem is that at this stage, if one fails the other is not able to proceed afetr the union all task 🙁